BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Code Search and Idiomatic Snippet Synthesis
DTSTART:20160408T110000
DTEND:20160408T120000
DTSTAMP:20260407T103055Z
UID:d2adca08411ecfa24f196519d22707d43391ace1cdf6bf7b363515e6
CATEGORIES:Conferences - Seminars
DESCRIPTION:by Mukund RaghothamanAbstract\nIn recent years\, the program a
 nalysis and synthesis communities are realizing the value of large open-so
 urce code repositories such as GitHub and BitBucket. These repositories ca
 n greatly impact the field: from providing better real-world benchmarks fo
 r existing algorithms\, to facilitating entirely new techniques for code c
 ompletion and anomaly detection.\nIn this talk\, we will consider the prob
 lem of API exploration. Modern programming frameworks come with large libr
 aries\, with diverse applications such as for matching regular expressions
 \, parsing XML files and sending email. Programmers often use search engin
 es such as Google and Bing to learn about existing APIs. I will describe S
 WIM (Synthesize What I Mean)\, a tool which suggests code snippets given A
 PI-related natural language queries such as ``generate md5 hash code''. Th
 e query does not need to contain framework-specific trivia such as the typ
 e names or methods of interest.\nI will address three specific problems: i
 nferring ``idioms'' from large code repositories\, natural language proces
 sing to understand the input queries\, and the architecture of the SWIM sy
 nthesizer which allows fast response times and easy collaboration between 
 the NLP and programming language researchers.\nWe evaluated SWIM with 30 c
 ommon C# API-related queries received by Bing. For 70% of the queries\, th
 e first suggested snippet was a relevant solution\, and a relevant solutio
 n was present in the top 10 results for all benchmarked queries. The onlin
 e portion of the workflow is also very responsive\, at an average of 1.5 s
 econds per snippet.\nThis is joint work with Yi Wei and Youssef Hamadi dur
 ing a summer internship at Microsoft Research Cambridge.Bio\nMukund Raghot
 haman is a Ph.D. student at the University of Pennsylvania\, advised by Ra
 jeev Alur\, and funded by the NSF ExCAPE grant.\nHis research goal is to m
 ake programming easier by building new programming abstractions and assist
 ance tools. In his thesis\, he designs DReX\, a domain-specific language t
 o describe stream transformations. He are now studying extensions to quant
 itative functions\, approximate query evaluation\, and applications to the
  static analysis of string manipulating programs.\nMore broadly\, he is in
 terested in formal verification and program synthesis. Program synthesis i
 s the problem of converting human intentions into concrete programs. The i
 nput is often vague and exploratory: he spent two summers working with You
 ssef Hamadi and Yi Wei on the synthesis of idiomatic code snippets for the
  Bing Code Search Tool at Microsoft Research Cambridge. He was also part o
 f the team that formalized SyGuS.\nIn 2010\, he graduated from the Indian 
 Institute of Technology Guwahati with an undergraduate degree in computer 
 science.More information
LOCATION:BC 420 https://plan.epfl.ch/?room==BC%20420
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR